COMPUTED TOMOGRAPHY

Burden of waiting for surveillance CT colonographyin patients with screen-detected 6–9 mm polyps

Charlotte J. Tutein Nolthenius1,2 & Thierry N. Boellaard1& Margriet C. de Haan3

&

C. Yung Nio1 & Maarten G. J. Thomeer4 & Shandra Bipat1 &

Alexander D. Montauban van Swijndregt2 & Marie-Louise Essink-Bot6 &

Ernst J. Kuipers7,8 & Evelien Dekker5 & Jaap Stoker1

Received: 1 July 2015 /Revised: 25 January 2016 /Accepted: 26 January 2016 /Published online: 8 April 2016# The Author(s) 2016. This article is published with open access at Springerlink.com

AbstractPurpose We assessed the burden of waiting for surveillance CTcolonography (CTC) performed in patients having 6–9 mm co-lorectal polyps on primary screening CTC. Additionally, wecompared the burden of primary and surveillance CTC.Materials and methods In an invitational population-basedCTC screening trial, 101 persons were diagnosed with <3polyps 6–9 mm, for which surveillance CTC after 3 yearswas advised. Validated questionnaires regarding expectedand perceived burden (5-point Likert scales) were completedbefore and after index and surveillance CTC, also includingitems on burden of waiting for surveillance CTC. McNemar’stest was used for comparison after dichotomization.Results Seventy-eight (77 %) of 101 invitees underwent sur-veillance CTC, of which 66 (85%) completed the expected and62 (79 %) the perceived burden questionnaire. The majority ofparticipants (73 %) reported the experience of waiting for sur-veillance CTC as ‘never’ or ‘only sometimes’ burdensome.There was almost no difference in expected and perceived bur-den between surveillance and index CTC. Waiting for the re-

sults after the procedure was significantly more burdensome forsurveillance CTC than for index CTC (23 vs. 8 %; p=0.012).Conclusion Waiting for surveillance CTC after primary CTCscreening caused little or no burden for surveillance partici-pants. In general, the burden of surveillance and index CTCwere comparable.Key points• Waiting for surveillance CTC within a CRC screeningcaused little burden

• The vast majority never or only sometimes thought abouttheir polyp(s)

• In general, the burden of index and surveillance CTC werecomparable

• Awaiting results was more burdensome for surveillance thanfor index CTC

Keywords Colonography, computed tomographic/methods .

Mass screening/methods . Colorectal neoplasms/diagnosis .

Anxiety/epidemiology . Painmeasurement

* Charlotte J. Tutein Noltheniusc.j.tuteinnolthenius@amc.uva.nl

1 Department of Radiology, Academic Medical Center, University ofAmsterdam, PO Box 22700, 1100 DE Amsterdam, The Netherlands

2 Department of Radiology, Onze Lieve Vrouwe Gasthuis,Amsterdam, The Netherlands

3 Department of Radiology, Meander Medical Center,Amersfoort, The Netherlands

4 Department of Radiology, Erasmus University Medical Center,Rotterdam, The Netherlands

5 Gastroenterology and Hepatology, Academic Medical Center,University of Amsterdam, PO Box 22700, 1100DE Amsterdam, The Netherlands

6 Public Health, Academic Medical Center, University of Amsterdam,PO Box 22700, 1100 DE Amsterdam, The Netherlands

7 Gastroenterology and Hepatology, Erasmus University MedicalCenter, Rotterdam, The Netherlands

8 Internal medicine, Erasmus University Medical Center,Rotterdam, The Netherlands

Eur Radiol (2016) 26:4000–4010DOI 10.1007/s00330-016-4251-4

Introduction

Debate remains regarding the management of individuals with6–9 mm polyps at screening, and an option is surveillance CTcolonography, because of the low malignancy risk and slowgrowing rate of these polyps [1–4]. Successful surveillance withCTcolonography is amongst other factors dependent on patients’adherence, which is likely related to previous experiences withCT colonography and expectations toward the surveillance ex-amination. Furthermore, screening for colorectal cancer (CRC)may cause relevant side effects, such as discomfort, anxiety, anddistress. In particular, waiting for a follow-up examination withthe idea of having a lesion that may progress to cancer, andwaiting for the results of the surveillance examination may causeconsiderable negative psychological consequences. The conse-quences of awaiting surveillance CT colonography are currentlyunknown and have to be weighed against CTcolonography ben-efits [5]. Further, one can speculate about potential differences inexperience between first and second CT colonography, as pa-tients should know better what to expect the second time.

In this prospective study, we evaluated the burden ofwaiting 3 years for the procedure and the results of a surveil-lance CT colonography in patients having one or two small(6–9 mm) polyps on primary screening CT colonography. Wealso assessed the burden, pain, and embarrassment of surveil-lance CT colonography in comparison to index CTcolonography in our previously performed, randomized,population-based screening trial [6].

Materials and methods

Study design and participants

We enrolled asymptomatic 50–75 year adults with no prior orfamily history of CRC or adenomatous polyps, who underwentCT colonography in the context of an invitational population-based screening program (COCOS trial) [7, 8]. Eligible patientshad one or two 6–9 mm polyps that were prospectively identi-fied on screening CTcolonography, and were advised to under-go surveillance CT colonography after 3 years. At that time,management after surveillance CT colonography was not clear.

Ethics approval from the Dutch Health Council (2009/03WBO, The Hague, the Netherlands) was obtained for theCOCOS trial, in which the surveillance CT colonography wasincluded. Patients had already given their written informedconsent to be contacted for follow-up studies and consentedto this study.

Invitation

Invitations for surveillance CT colonography were sent bymail and patients were asked to contact one of the researchers

by phone. In the case of no response, a reminder letter wassent after 4 weeks and if no response to the reminder letter wasreceived, the person was contacted by phone once after anoth-er 4 weeks. Patients who underwent a CT colonography orcolonoscopy in the time between the index CT colonographyand the advised surveillance CTcolonography were excluded.Study information was sent to all included patients willing toparticipate. Non-participation reasons were summarized [9].

Procedure

CTcolonography preparation, colonic distention, and scan pro-tocol were identical to the index CTcolonography examination(see for detailed description Appendix I). Briefly, patientsunderwent a non-cathartic preparation the day before the exam-ination and 1.5 hours prior to the examination (total of 3x50mLof iodinated contrast agent; Telebrix Gastro, Guerbet,Aulnaysous-Bois, France) in combination with a low-fibre dietfor 1 day [10, 11]. Before colonic insufflation, 20 mg intrave-nous hyoscine butylbromide (Buscopan; Boehringer-Ingelheim, Ingelheim, Germany) was administered, after whichcolonic distention was achieved with automated pressure-controlled carbon dioxide insufflation (PROTOCO2L,Bracco, EZEM, Lake Success, NY, USA) through a thin, flex-ible rectal tube. When distension was considered sufficient viathe scout image, a breath-hold supine and prone CT was per-formed using a 64 multi-detector-row CT scanner (Brilliance,Philips Healthcare, Best, the Netherlands; SOMATOMSensation, Siemens Medical Solutions, Erlangen, Germany).Patients received no sedation or analgesics. Expert readers eval-uated all CTcolonographies within 2 weeks after the procedure.Patients were informed by phone about the results within2 weeks after the surveillance CT colonography. Colonoscopywas done if a lesion ≥6 mm was reported.

Questionnaires

The expected burden questionnaire (EBQ) and perceived bur-den questionnaire (PBQ) were identical to those used in indexCT colonography and are based on previous Dutch faecal oc-cult blood test (FOBT) screening pilots, and on studies investi-gating the acceptance of CT colonography and patient percep-tion of diagnostic tests for faecal incontinence [12–15]. For allincluded EBQs and PBQs we retrieved the correspondingEBQs and PBQs of index CT colonography, and the questionsfor index and surveillance CT colonography were compared.

Expected burden questionnaire (EBQ)

A pre-procedural questionnaire was sent to the participants toassess the expectations towards the CT colonography within4 weeks before the procedure. They were asked to completethe questionnaire prior to the screening procedure and to

Eur Radiol (2016) 26:4000–4010 4001

return it bymail in a prepaid envelope (whichwas identical forindex CTcolonography). The EBQ contained items regardingexpected embarrassment, pain, and burden of the bowel prep-aration and the examination itself. All but two items werescored on 5-point Likert scales (1 =not at all; 2 = slightly;3= some; 4= rather; 5= extremely) [16]. Two items regardingthoughts in the past 3 years were scored on 4-point Likertscales (1 = never; 2 = sometimes; 3 = regularly; 4 = often).Expected burden questionnaires completed after surveillanceCT colonography were excluded from analysis.

Perceived burden questionnaire (PBQ)

The perceived burden, pain, and embarrassment of CTcolonography and different procedural aspects were assessedusing a post-procedural questionnaire, sent by mail after theparticipants had been informed about the CT colonography re-sult. The PBQ contained items regarding the perceived embar-rassment, pain, and burden of the bowel preparation, the exam-ination itself, and the overall burden of the CT colonographyexamination (same 5-point Likert scale as pre-procedural [16]).We also collected information on the willingness to return forfuture screening rounds. PBQs that were completed after12weeks from the surveillance CTcolonographywere excludedfrom analysis, as we considered this time interval too long for anaccurate representation of the perceived burden (for index CTcolonography, this was 6 weeks)

Surveillance-specific questions

In addition to the EBQ for the index CT colonography, itemsregarding the burden and fear of waiting for (the results of)surveillance CT colonography were included. All but twoitems were scored on 5-point Likert scales. Two items regard-ing thoughts in the past 3 years were scored on 4-point Likertscales (1=never; 2= sometimes; 3= regularly; 4=often). ThePBQ included additional items on the most burdensome as-pect of the surveillance CTcolonography (waiting for surveil-lance CT colonography, preparation, examination, abdominalsymptoms afterward, or waiting for the results) and of theentire CT colonography screening procedure ((results of) in-dex or surveillance CT colonography).

Impact of event scale (IES)

The level of colorectal cancer (CRC)-specific distress wasassessed using the Impact of Event Scale (IES) sent togetherwith the EBQ (IES-I) and PBQ (IES-II) [17–19]. The 15 IESitems were tailored to the specific Bevent^: Ba diagnosis of co-lorectal cancer .̂ All items were scored on 4-point Likert scalesmeasuring whether this was present (0=not at all; 1= rarely;3=sometimes; 5=often) during the past 7 days. The summaryscore of the total scale could range from 0 to 75, with a higher

score indicatingmore cancer-specific distress (0–8: nomeaning-ful impact; 9–25: impact event; 26–43 powerful impact event;44–75 severe impact event). The IESwas found to be sufficient-ly reliable (Cronbach α for IES-I and IES-II was 0.87 and 0.89,respectively) to allow these analyses to be performed.

Statistical analysis

Demographics and items specific for surveillance CTcolonography were described. Corresponding questions onEBQs and PBQs were compared between index and surveil-lance CT colonography, and only paired data were used.Answers to questions with 4-point and 5-point ordinal scaleswere first dichotomized (totally agree/agree versus disagree/totally disagree and not at all/slightly versus some/rather/ex-tremely) and differences were assessed using the McNemar testfor paired proportions. One question regarding expectations to-wards surveillance CTcolonography required a different dichot-omization because of different response categories (better/slightly better/as expected versus slightly worse/worse than ex-pected). We used Mann-Whitney U tests to test the significanceof differences in the mean IES scores by sex and surveillanceCT colonography results. To test for differences in IES overtime, paired Wilcoxon signed-rank tests were conducted. A p-value of <0.05 was considered significant. To determine theclinical relevance of the significant difference between means,we used the minimal important difference (MID), which is de-fined as half of a standard deviation. SPSS V.20.0 for Windows(SPSS, Chicago, Ill) was used to perform all statistical tests.

Results

Response and respondent characteristics

Between July 2012 and May 2014, 101 patients were invitedto undergo surveillance CT colonography (Fig. 1). Twenty-three (23%) of the 101 invitees were excluded for surveillanceCT colonography (four passed away; ten underwent a colo-noscopy examination prior to the invitation of the surveillanceCT colonography, of which five were at the patient’s ownrequest (Appendix II); nine did not wish to participate, themost common reason mentioned was other health problems).

Seventy-eight patients (77 %) underwent a surveillanceCT colonography (median age 63.5 IQR [60.0–72.0]; 53 %male) at a mean surveillance interval of 3.3 years (SD 0.3;range 3.0–4.6 years) (Table 1). The EBQ was completed intime by 66 (85 %) surveillance participants, the IES-I in 65(83 %) participants. The PBQ and IES-II were returnedwithin 12 weeks of surveillance CT colonography by 62patients (79 %). Sixty-four EBQs and 48 PBQs of indexand surveillance CT colonography were available for ahead-to-head comparison.

4002 Eur Radiol (2016) 26:4000–4010

Waiting for the surveillance CT colonography

The vast majority of the respondents never or only sometimesthought about the colonic polyp(s) (89 %; 57/64) or the recom-mended surveillance CT colonography (84 %; 54/64) (Fig. 2).Thoughts about the presence of the colonic polyp(s) or thewaiting for surveillance CTcolonography were for the majorityof participants not at all or only slightly burdensome (75 %; 47/63 and 73 %; 46/63, respectively). Waiting for surveillance CTcolonography did not result in anxiety or slight anxiety in 86 %(54/63). The vast majority expected the waiting for the resultsto be not at all or only slightly burdensome (70 %; 45/64) orexpected to be not at all or only slightly anxious (72 %; 46/64)while waiting for the results of surveillance CT colonography.

Expected burden in comparison to index CTcolonography

Expected burden for surveillance CT colonography was ratedsimilarly as for index CT colonography, with the exceptionthat the bowel preparation was expected to be not or onlyslightly embarrassing by a larger proportion of participantsfor surveillance CT than for index CT (87 %; 55/63 versus71 %; 45/63, respectively; p=0.031) (Fig. 3). The proportionof participants who thought that the evaluation of extracolonicstructures on CT colonography was advantageous was largerin surveillance than in index CT colonography (98 %; 63/64versus 89 %; 57/64; p=0.031) (Appendix III).

Perceived burden in comparison to index CTcolonography

Items on perceived burden of bowel preparation, CTcolonography examination, abdominal complaints in the weeksof the examination, and the entire screening procedure were

Fig. 1 Overview of response tothe expected and perceivedburden questionnaire (EBQ andPBQ) including Impact of EventScale (IES) among 101surveillance CT colonographyparticipants. aFour passed away,ten underwent a colonoscopyprior to invitation for surveillanceCT colonography, and nine didnot wish to participate for variousreasons. bCompleted the EBQ &IES-I and PBQ & IES-II in time(prior to the surveillance CTcolonography and within12 weeks of surveillance CTcolonography, respectively)

Table 1 Baseline characteristics of CT colonography surveillanceparticipants

Participants in surveillance CTcolonography (n = 78)

Age in years (median, IQR) 64 (60–72)

Gender (% male) 41 (53 %)

Married/lived together (%)a,b 59 (81 %)

Socioeconomic status (mean, SD)a,b,c 2.8 (SD 1.4)

Educationa,b

- Elementary (%) 5 (7 %)

- Secondary (%) 45 (62 %)

- Tertiary and postgraduate (%) 23 (32 %)

Previous experience (%)b

- CT colonography 1 (2 %)

- Colonoscopy 4 (6 %)

a At time of index CT colonographybAs not all respondents completed the questions on their marital status,education, and prior endoscopy experience, the percentages mentionedfor these items are not based on the total number of respondents, but onthe total number of participants who answered those questionsc Socioeconomic status was categorised as very low, low, medium, high,and very high (1–5)

Eur Radiol (2016) 26:4000–4010 4003

rated comparably for index CT colonography, with the CTcolonography examination itself cited as the most burdensomepart and insufflation of CO2 as the most burdensome/painfulpart of the CT examination (Appendix IV). Two items differed:waiting for the results was considered some, rather, or extremelyburdensome by 23 % (11/47) in surveillance CT colonographyversus 6 % (3/47) of participants in index CT colonography(p=0.021) (Fig. 4). The surveillance CT colonography was ex-perienced as expected by 31% (15/48) and worse than expectedby 6 (3/48), versus 15 % (7/48) and 23 % (11/48), respectively,in index CT colonography (p=0.011).

Evaluation of entire screening procedure

The most burdensome part of the surveillance CTcolonography procedure was considered to be the CT exami-nation itself (33 %; 16/48) followed by bowel complaints(25 %; 12/48) and waiting for the results (25 %; 12/48). Forindex CT colonography, the bowel preparation was indicatedas the most burdensome part (35 %; 17/48), followed by theexamination itself (29 %; 14/48) and the bowel complaints(25 %; 12/48). Waiting 3 years for surveillance CTcolonography was indicated as the most burdensome part byonly 8 % (4/48) of invitees. There was no significant differ-ence between surveillance and index CTcolonography for the

proportion of participants who would recommend CTcolonography in screening (96 %; 46/48 versus 98 %; 47/48;p = 1.000) and who would participate in future screeningrounds (98 %; 46/47 versus 94 %; 44/47; p=0.625).

Psychological distress

Fifty-five paired IES scores were available for analysis. Thetotal IES scores were significantly higher for women than formen prior to and after surveillance CT colonography (9.4 (SD9.8) and 3.6 (SD 5.3); p=0.006 and 11.9 (SD 11.9) and 4.3 (SD4.7); p=0.012, respectively) (Fig. 5a). Differences were largerthan the MID and therefore considered to be clinically relevant.

The total IES score of the group of patients with a positiveCT colonography result significantly increased over time(from 5.2 SD (6.6) to 8.2 (SD 10.1); p = 0.006) (Fig. 5b).The difference was smaller than the MID and therefore notconsidered to be clinically relevant.

Discussion

This study demonstrated that the knowledge of having a 6–9 mm colonic polyp, and subsequently, being under surveil-lance caused little or no burden. It furthermore showed that

Fig. 2 Thoughts on colonicpolyps and experienced burden inthe period between the index andsurveillance CT colonography(the past 3 years). Anxiety andexpected burden while waiting for(the results of) surveillance CTcolonography

4004 Eur Radiol (2016) 26:4000–4010

waiting for surveillance CTcolonography in screening for CRCcaused little to no anxiety or burden in the vast majority ofparticipants. In general, expectations towards and perceived

burden of surveillance CT colonography were similar to indexCT colonography 3 years before. Having to wait for the CTcolonography results after the surveillance CT colonography

Fig. 3 Reluctance to undergoindex (screening) CTcolonography and surveillanceCT colonography, and expectedembarrassment, pain, and burdenof bowel preparation and CTcolonography of bothexaminations. P values ofMcNemar’s test afterdichotomization are presented ontop of the bars

Fig. 4 Perceived burden of theentire CT colonographyexamination, waiting for the CTcolonography results and entirescreening procedure withsurveillance CT colonography, incomparison to index CTcolonography. P-values ofMcNemar’s test afterdichotomization are presented ontop of the bars

Eur Radiol (2016) 26:4000–4010 4005

was the only factor considered significantly more burdensomethan for index CT colonography (23 versus 8 %). SurveillanceCT colonography turned out to be worse than expected com-pared to index CTcolonography far less often, in 6 versus 23%of patients, respectively. Surveillance for CRC with CTcolonography appeared to have no major impact on most pa-tients; however, women did experience significantly more dis-tress than men. The large majority of participants would rec-ommend participating in CT colonography for CRC screening(96 %) and would participate again in the future (98 %).

To the best of our knowledge, our study is the first toevaluate the burden of waiting for surveillance CTcolonography in patients with one or two 6–9 mm polypsdetected at primary screening CT colonography. Our studydesign with identical questionnaires made a head-to-headcomparison between expected and perceived burden of indexand surveillance CT colonography possible. Both question-naires had been validated in previous CRC screening pilots.

Patients did not seem to worry too much about the possiblepresence of colonic polyp(s), or did not find the knowledge ofhaving this/these polyp(s) burdensome, which also can be con-cluded from the small fraction of patients (8%) who consideredwaiting for surveillance CTcolonography themost burdensomeaspect of the entire screening procedure. One must realize thatour study population does not represent a random selectedgroup of persons. Individuals chose to participate in the initialCRC screening trial and, in exchange for being informed abouttheir health, accepted certain consequences—for example, (theburden of) surveillance CT colonography. In addition, ten ex-cluded patients underwent a preliminary colonoscopy, of whichfive were at their own request. In these patients, it seems likelythat anxiety or worry was a barrier for waiting 3 years forsurveillance CT colonography and that they were probablymore prone to psychological stress.

In addition to the selection bias described noted above, anumber of other potential limitations should be acknowledged.The surveillance population originated from a primary screeningtrial (COCOS trial), which was powered for comparing the par-ticipation rate and yield of CTcolonography versus colonoscopyscreening and not for evaluating burden in surveillance CTcolonography, therefore leading to a relatively small surveillancepopulation [10]. In addition, we only used paired data for solidcomparison analyses, resulting in the exclusion ofmore question-naires. Because of those limited number of patients, we chose arelatively wide interval (12 weeks) in which the PBQ could bereturned, to retain as many questionnaires for analyses as possi-ble. This possibly could have affected patients’ evaluations. Anassessment of PBQ and cancer-specific distress (IES) before re-ceiving the results of surveillance CT colonography might haveshowed higher burden scores and/or distress, as waiting for theresults was reported as burdensome in 25 %. Also, an extraassessment later in time (after the results of the colonoscopy)could have provided us with information about the duration of

higher distress experienced by certain patients [20]. Because ofthe associated questions (e.g. three questions on bowel prepara-tion, for pain, burden, and embarrassment), there is a chance ofeventually finding a significant difference on one question/out-come, aswas probably the casewith the expected embarrassmentabout bowel preparation given the comparable expectations onall other items.

Waiting for the results of a (screening) study is burdensome[5]. The significantly higher proportion of patients that con-sider waiting for the results of surveillance CT colonographyto be somewhat, rather, or extremely burdensome in compar-ison to index CT colonography (23 versus 8 %) is most likelyrelated to the knowledge of having a colonic polyp that mayhave grown. However, the waiting time for the results of thesurveillance CT colonography in our study was unusuallylong (2 weeks), as the CTwas evaluated by multiple observers

Fig. 5 CRC-specific distress score (Impact of Event Scale [IES] total)over time. a Total IES scores for the total group, separated by gender.Total IES scores in women were significantly higher than for men prior toand after surveillance CT colonography (p = 0.006 and p = 0.012,respectively). b Total IES scores for the total group, separated by resultof the surveillance CT colonography. Total IES score of the group with apositive CTcolonography result increased (p = 0.006). Total IES scores ofthe group with a negative result decreased (p = 0.655)

4006 Eur Radiol (2016) 26:4000–4010

within the framework of a different study. In daily practice, thewait will be shorter and this might decrease the burden. Atcolonoscopy, patients are directly informed about the results,which is very much appreciated by patients [6].

The significantly larger proportion of patients reportingsurveillance CT colonography to be better than or as expectedcompared to index CT colonography (94 versus 77 %) was inaccordance with our hypothesis that patients previously hadunderestimated certain elements of index CT colonography[6]. As the study information and guidance during the CTcolonography was identical to index CT colonography, themore realistic expectations are most likely due to previousexperience with CT colonography. This was, however, notaccompanied by a higher expected or lower perceived burdenin surveillance CT colonography. The CT colonography ex-amination itself was identified as the most burdensome part ofthe entire screening procedure when participants were forcedto choose one aspect alone, and in line with previous studies,the colonic insufflation with CO2 was considered the mostburdensome and painful part of CT colonography [6, 21].Although analgesia has shown to reduce total procedural painand burden in CT colonography [22], this use of it remains amatter of practical hurdles, side effects, and costs.

The Impact of Event Scale (IES) has been used in thepsycho-oncology literature as a measure of cancer-relatedanxiety, amongst other types in screening populations [5, 23,24]. Althoughwe noticed a significant increase in the total IESscore in patients with a positive result, we did not notice anyclinically relevant changes in cancer-specific distress, whichwas in accordance with other screening studies [5, 23]. Thehigher distress scores in women than in men were also report-ed in other studies [5, 25, 26]. This should be considered whensupplying information and providing counselling for CRCscreening.

In conclusion, waiting for surveillance CT colonographycaused little discomfort. In general, the expected and per-ceived burden of surveillance and index CT colonographywere comparable. Since waiting for the results for surveillanceCT colonography proved to be burdensome, minimizing thewaiting time for the test result is recommended. It is alsoreassuring that after surveillance CT colonography, almostevery patient would participate again in future screeningrounds.

Acknowledgments The scientific guarantor of this publication is J.Stoker, MD, PhD. This study was funded by the Dutch Cancer Society(KWF Kanker Bestrijding 2012–5698). Philips Healthcare (Best, theNetherlands) provided the CT colonography workstations. Neither wereinvolved in designing or conducting this study, had access to the data, orwere involved in data analysis or preparation of this manuscript. Oneauthor (J Stoker) of this manuscript declares relationships with the fol-lowing companies: Robarts. One of the authors has significant statisticalexpertise. Institutional Review Board approval was obtained. Writteninformed consent was obtained from all subjects (patients) in this study.

Some study subjects or cohorts have been previously reported in: deWijkerslooth et al. Gut 2012 ‘Burden of colonoscopy compared to non-cathartic CT-colonography in a colorectal cancer screening programme:randomised controlled trial. Methodology: prospective, cross sectionalstudy, multicenter study.

We thank research nurses Isha Verkaik and Laurens Groenendijk fortheir research support.

Compliance with ethical standards All procedures performed in stud-ies involving human participants were in accordance with the ethicalstandards of the institutional and/or national research committee and withthe 1964 Helsinki declaration and its later amendments or comparableethical standards.

Appendix I Detailed description of CT colonography procedureA non-cathartic preparation consisting of two 50-mL doses of iodin-

ated contrast agent (Telebrix Gastro, Guerbet, Aulnaysous-Bois, France)was given on the day before the examination, and another 50 mL wasgiven 1.5 hours prior to the examination (total = 150 mL) in combinationwith a low-fibre diet for 1 day [8, 9].

All CT colonography examinations were performed by experiencedradiologic technologists. Both supine- and prone-position CT images wereobtained on two 64-slice CT-scanners (Brilliance, Philips Healthcare, Best,the Netherlands; SOMATOM Sensation, Siemens Medical Solutions,Erlangen, Germany) using a low-dose protocol; with collimation64× 0.625 mm, slice thickness 0.9 mm, reconstruction interval 0.7 mm,tube voltage 120 kV, and 25 reference mAs (for Brilliance) and collimation128× 0.6 mm, slice thickness 1.0 mm, reconstruction interval 0.7 mm, tubevoltage 120 kV, and 16 ref mAs (for SOMATOM Sensation). Before co-lonic insufflation, 20 mg intravenous hyoscine butylbromide was adminis-tered (if contraindicated, 1 mg of glucagon hydrochloride was used intra-venously). Distention of the colon was obtained via a thin, flexible rectalcatheter using automatic insufflation of carbon dioxide (PROTOCO2L,Bracco, EZEM, Lake Success, NY, USA).

All CT colonographies were evaluated within 2 weeks after theprocedure by an experienced abdominal radiologist using primary2D (window setting 1500, -250 HU) with 3D problem solving(Amsterdam: View Forum, Philips, Best, the Netherlands), and bytwo o f f ou r t r a i n ed t e chno log i s t s ( expe r i enc e o f a l lfour = approximately ±1200 CT colonographies), read using primary2D and followed by 3D [5]. All readers finished with computer-aided detection (CAD) evaluation. Patients were informed aboutthe results within 2 weeks of the CT colonography. Colonoscopywas done if at least one of the observers reported a lesion ≥6 mm.

Appendix II Reasons for early surveillance CT colonography orcolonoscopy instead of the advised surveillance CTcolonography at 3 years

Table 2 Reasons for early surveillance CTcolonography or colonoscopyinstead of the advised surveillance CT colonography at 3 years

Number of patients

Surveillance CT colonography at 1.5 years

Preference directly after initial CTcolonography at screening trial 2

Early colonoscopy

Preference directly after initial CTcolonography at screening trial 2

Abdominal complaints 4

Worries about the presence of polyps 1

Refused to give information 1

Total 10

Eur Radiol (2016) 26:4000–4010 4007

Appendix III General statements on bowel cancer screening inindex and surveillance CT colonography. P-values of McNemar’s testafter dichotomization are presented on top of the bars

I have a largerchance on

bowel cancer

Bowel canceris a seriouscondition

I fear theresults

0

p=0.424

Evaluationextracolonicstructures isadvantageous

There are moredisadvantages

thanadvantages

p=1.000 # p=0.118 p=0.031 p=1.000

Totally agree

Agree

Disagree

Totally disagree

Left bar: index CTCRight bar: surveillance CTC

100%

Screening forbowel cancer

is useful

80%

60%

40%

20%

4008 Eur Radiol (2016) 26:4000–4010

Fig. 6 General statements on bowel cancer screening in index andsurveillance CT colonography. P-values of McNemar’s test after dichot-omization are presented on top of the bars

Appendix IV Perceived burden of bowel preparation and differentaspects of surveillance CT colonography in comparison to index CTcolonography. P-values of McNemar’s test after dichotomization arepresented on top of the bars

Open Access This article is distributed under the terms of the CreativeCommons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits anynoncommercial use, distribution, and reproduction in any medium,provided you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons license, and indicate ifchanges were made.

References

1. Lieberman D, Moravec M, Holub J, Michaels L, Eisen G (2008)Polyp size and advanced histology in patients undergoing colonos-copy screening: impl ica t ions for CT colonography.Gastroenterology 135:1100–1105

2. Kim DH, Pickhardt PJ, Taylor AJ et al (2007) CT colonographyversus colonoscopy for the detection of advanced neoplasia. N EnglJ Med 357:1403–1412

3. Spada C, Stoker J, Alarcon O et al (2015) Clinical indications forcomputed tomographic colonography: european society of

gastrointestinal endoscopy (ESGE) and european society of gastro-intestinal and abdominal radiology (ESGAR) guideline. Eur Radiol25:331–345

4. Tutein Nolthenius CJ, Boellaard TN, de Haan MC et al (2015)Computer tomography colonography participation and yield in pa-tients under surveillance for 6-9 mm polyps in a population-basedscreening trial. Eur Radiol

5. van den Bergh KA, Essink-Bot ML, Bunge EM et al (2008) Impactof computed tomography screening for lung cancer on participants ina randomized controlled trial (NELSON trial). Cancer 113:396–404

6. deWijkerslooth TR, de HaanMC, Stoop EM et al (2012) Burden ofcolonoscopy compared to non-cathartic CT-colonography in a co-lorectal cancer screening programme: randomised controlled trial.Gut 61:1552–1559

7. de Haan MC, Nio CY, Thomeer M et al (2012) Comparing thediagnostic yields of technologists and radiologists in an invitationalcolorectal cancer screening program performed with CTcolonography. Radiology 264:771–778

8. de Wijkerslooth TR, de Haan MC, Stoop EM et al (2010) Studyprotocol: population screening for colorectal cancer by

Embarrassment Pain Burdensome

Bowel preparation, consequences of iodinatedcontrast medium

Not at all

Slight

Some

Rather

Extremely

Left bar: index CTCRight bar: surveillance CTC

p=0.375 p=0.687 p=0.754

60%

80%

100%

0

40%

20%

Injection ofbuscopan

Pain Burdensome

Insufflation of CO2

p=1.000 p=1.000 p=0.791 p=0.344

Placing therectal

catheter

p=1.000

Turningaround onCT table

Eur Radiol (2016) 26:4000–4010 4009

Fig. 7 Perceived burden of bowel preparation and different aspects ofsurveillance CT colonography in comparison to index CT colonography.

P-values of McNemar’s test after dichotomization are presented on top ofthe bars

colonoscopy or CT colonography: a randomized controlled trial.BMC Gastroenterol 10:47

9. van Rijn AF, van Rossum LG, DeutekomM et al (2008) Low prioritymain reason not to participate in a colorectal cancer screening programwith a faecal occult blood test. J Public Health (Oxf) 30:461–465

10. Stoop EM, de Haan MC, de Wijkerslooth TR et al (2012)Participation and yield of colonoscopy versus non-cathartic CTcolonography in population-based screening for colorectal cancer:a randomised controlled trial. Lancet Oncol 13:55–64

11. Liedenbaum MH, Denters MJ, Zijta FM et al (2011) Reducing theoral contrast dose in CTcolonography: evaluation of faecal taggingquality and patient acceptance. Clin Radiol 66:30–37

12. Denters MJ, Deutekom M, Fockens P, Bossuyt PMM, Dekker E(2009) Implementation of population screening for colorectal can-cer by repeated fecal occult blood test in the Netherlands. BmcGastroenterol 9

13. van Gelder RE, Birnie E, Florie J et al (2004) CTcolonography andcolonoscopy: assessment of patient preference in a 5-week follow-up study. Radiology 233:328–337

14. Deutekom M, Terra MP, Dijkgraaf MGW et al (2006) Patients’perception of tests in the assessment of faecal incontinence. Br JRadiol 79:94–100

15. Hol L, van Leerdam ME, van Ballegooijen M et al (2010)Screening for colorectal cancer: randomised trial comparingguaiac-based and immunochemical faecal occult blood testingand flexible sigmoidoscopy. Gut 59:62–68

16. Likert R (1932) A technique for the measurement of attitudes. York,New

17. van der Ploeg E, Mooren TT, Kleber RJ, van der Velden PG, BromD (2004) Construct validation of the Dutch version of the impact ofevent scale. Psychol Assess 16:16–26

18. Horowitz M,Wilner N, AlvarezW (1979) Impact of Event Scale: ameasure of subjective stress. Psychosom Med 41:209–218

19. BromD, Kleber RJ (1985) De Schok Verwerkings Lijst [The Dutchversion of the Impact of Event Scale]. Ned Tijdschr Psychol 40:164–168

20. Lowe JB, Balanda KP, Del Mar C, Hawes E (1999) Psychologicdistress in women with abnormal findings in mass mammographyscreening. Cancer 85:1114–1118

21. Jensch S, Bipat S, Peringa J et al (2010) CT colonographywith limited bowel preparation: prospective assessment ofpatient experience and preference in comparison to opticalcolonoscopy with cathartic bowel preparation. Eur Radiol20:146–156

22. Boellaard TN, van der Paardt MP, Hollmann MW et al (2013) Amulti-centre randomised double-blind placebo-controlled trial toevaluate the value of a single bolus intravenous alfentanil in CTcolonography. BMC Gastroenterol 13:94

23. Cullen J, Schwartz MD, Lawrence WF, Selby JV, Mandelblatt JS(2004) Short-term impact of cancer prevention and screening activ-ities on quality of life. J Clin Oncol 22:943–952

24. Taylor KL, Shelby R, Gelmann E,McGuire C (2004) Quality of lifeand trial adherence among participants in the prostate, lung, colo-rectal, and ovarian cancer screening trial. J Natl Cancer Inst 96:1083–1094

25. Kruijshaar ME, Kerkhof M, Siersema PD et al (2006) The burdenof upper gastrointestinal endoscopy in patients with Barrett's esoph-agus. Endoscopy 38:873–878

26. Thiis-Evensen E, Wilhelmsen I, Hoff GS, Blomhoff S, SauarJ (1999) The psychologic effect of attending a screeningprogram for colorectal polyps. Scand J Gastroenterol 34:103–109

4010 Eur Radiol (2016) 26:4000–4010